Progress in understanding the evolution of variation at the MHC has been slowed by an inability to assess the relative roles of mutation vs. intragenic recombination in contributing to observed polymorphism. Recent theoretical advances now permit a quantitative treatment of the problem, with the result that the amount of recombination is at least an order of magnitude greater than that of mutation in the history of class II genes. We suggest that this insight allows progress in evaluating the importance of other factors affecting the evolution of the MHC. We investigated the evolution of MHC class II Eβ sequence diversity in the genus Peromyscus. We find evidence for extensive recombination in the history of these sequences. Nevertheless, it appears that intragenic recombination alone is insufficient to account for evolution of MHC diversity in Peromyscus. Significant differences in silent variation among subgenera arose over a relatively short period of time, with little subsequent change. We argue that these observations are consistent with the effects of historical population bottleneck(s). Population restrictions may explain general features of MHC evolution, including the large amount of recombination in the history of MHC genes, because intragenic recombination may efficiently regenerate allelic polymorphism following a population constriction.